Hi,
I’m thinking about making a linear power supply for my Sony VFet n-channel amp.
Wondering what approach folks have taken when making one?
Curious to see what experiences have been when adding one as far as the Amps performance changed from using a linear power supply vs using the SMPS.
Thanks!
Per
I’m thinking about making a linear power supply for my Sony VFet n-channel amp.
Wondering what approach folks have taken when making one?
Curious to see what experiences have been when adding one as far as the Amps performance changed from using a linear power supply vs using the SMPS.
Thanks!
Per
If building a single supply to power both channels, I would use a 300VA 30VAC power transformer feeding a CLC filter. The CLC would be 22,000uF - Hammond 159ZJ - 22,000uF. Perhaps then add a RC stage to each channel, say 0.1R - 1,000uF. Or instead of 1,000uF electrolytic, substitute a lower capacitance motor run capacitor (film).
Yes, all my SIT amps (no diyAudio Sony VFET) have CLC power supplies. They are all mono-blocks so they do not have shared power supplies so they do not have a RC section after the CLC.
By the way the 30VAC secondary was an estimate. The final voltage may be a bit high, but a 28VAC transformer may or may not be a bit low.
I just realized that I mentioned 22,000uF for the C, but in my amps I actually used 2 x 22,000uF for each C.
By the way the 30VAC secondary was an estimate. The final voltage may be a bit high, but a 28VAC transformer may or may not be a bit low.
I just realized that I mentioned 22,000uF for the C, but in my amps I actually used 2 x 22,000uF for each C.
Last edited:
Thank you!
I find the VFet to be quite “vivid” in its stock presentation with the SMPS. Comparing it to my Hiraga Super 30, it presents as being more on the analytical side.
Your observation gives me something to consider as I move forward. I was hoping to add a bit more drive to the bottom end of the sound with with a linear power supply without losing the sense of air and space tue amp already has. Based on my very limited experience, I would like to think a CLC would do this?
Per
I find the VFet to be quite “vivid” in its stock presentation with the SMPS. Comparing it to my Hiraga Super 30, it presents as being more on the analytical side.
Your observation gives me something to consider as I move forward. I was hoping to add a bit more drive to the bottom end of the sound with with a linear power supply without losing the sense of air and space tue amp already has. Based on my very limited experience, I would like to think a CLC would do this?
Per
As far as bottom end... I send everything under 80Hz to a really really good powered subwoofer (servo, made for music) because I think there are about 3,874 good reasons to do so. And let these (high passed) low power Class A amps do the rest, taking advantage of those strengths... again IMO
There is a whole thread devoted to building the original VFET lottery amp. The techniques for the P-type generally work for the N-type as well.
One thing to try first is adding some extra capacitance to the power rail right at each PCB. I started with 1000 uF, 50V caps at first, while other builders had good results with 2200 uF. The fact that the second version (N-type) moved to a larger filter board inside the amp justified this approach. Mark Johnson has designed a more extensive filter board for use in the N-type amp.
After I built the external linear supply, I added 30,000 uF of total capacitance to each channel inside the amp chassis. The CapMx style supply filters benefit from having bulk capacitance added to their output.
Each stage of modification had the effect of noticeably improving bass response and definition. There was also improvement in overall depth, separation and detail throughout the midrange. My impression of the DIY VFET amp is that it performs well above its weight class once more attention is paid to giving it an excellent power supply.
One thing to try first is adding some extra capacitance to the power rail right at each PCB. I started with 1000 uF, 50V caps at first, while other builders had good results with 2200 uF. The fact that the second version (N-type) moved to a larger filter board inside the amp justified this approach. Mark Johnson has designed a more extensive filter board for use in the N-type amp.
After I built the external linear supply, I added 30,000 uF of total capacitance to each channel inside the amp chassis. The CapMx style supply filters benefit from having bulk capacitance added to their output.
Each stage of modification had the effect of noticeably improving bass response and definition. There was also improvement in overall depth, separation and detail throughout the midrange. My impression of the DIY VFET amp is that it performs well above its weight class once more attention is paid to giving it an excellent power supply.
Totally understand and get that approach! As so many of us likely do, I have some “options” when it comes to speakers I have to listen to - ranging from small full range speakers to big box 3 way horns. What I don’t have, or make use of are subs. One day perhaps, but for now, I just want to optimize the amp.
thank you!
Started reading through the thread - wish there was a way to print out an entire thread. Would help so much when trying to isolate what is being researched as well as make notes.
I’m a paper/pencil user
What you wrote, is exactly what I was hoping to have confirmed!
Per
